Next Article in Journal
Predicting Methane Concentrations in Underground Coal Mining Using a Multi-Layer Perceptron Neural Network Based on Mine Gas Monitoring Data
Previous Article in Journal
The Effect of Environmental Damage Costs on the Performance of Insurance Companies
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 19
10.3390/su16198390
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Carbon Reduction Effects in Transport Infrastructure: The Mediating Roles of Collusive Behavior and Digital Control Technologies

by
Da Wang
,
Chongsen Ma
*,
Yun Chen
,
Ai Wen
,
Mengjun Hu
and
Qi Luo
School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha 410205, China
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(19), 8390; https://doi.org/10.3390/su16198390
Submission received: 10 September 2024 / Revised: 24 September 2024 / Accepted: 25 September 2024 / Published: 26 September 2024
(This article belongs to the Section Sustainable Transportation)
Browse Figures
Versions Notes

Abstract

:
Many countries have committed to carbon reductions and carbon neutrality targets in response to the Paris Agreement and Sustainable Development Goals (SDGs). With economic development, the transportation sector has become a major source of carbon emissions. In China, transport infrastructure—as an important carrier of the transportation sector—is important for controlling carbon emissions from this sector and achieving carbon neutrality and the targets of the SDGs. However, most studies have focused on transport vehicles and neglected transport infrastructure. Furthermore, the influences of collusive behavior and digital control technologies on the carbon reduction process have not yet been examined. This study aimed to analyze the influencing factors in the carbon reduction process in transport infrastructure. This study uses partial least squares structural equation modeling (PLS-SEM) to analyze the factors influencing carbon reductions in transport infrastructure and the mediating roles of collusive behavior and digital control technologies in the carbon reduction process. Low-carbon technologies, digital control technologies, and collusive behavior have positive direct and indirect effects on the carbon reduction effect. Digital control technologies have a positive effect on low-carbon regimes. Low-carbon technologies influence carbon reduction effects. Collusive behavior plays a mediating role in low-carbon regimes. Finally, the industrial structure influences carbon reduction effects. This study extends China’s carbon emission research in the transportation sector by focusing on infrastructure rather than vehicles. Additionally, this is the first study to incorporate collusive behavior and digital control technologies into the framework to analyze the impact of carbon reductions. The study also employs PLS-SEM to explore effective carbon reduction paths. The findings provide decision-making support for controlling carbon reductions in transport infrastructure.

1. Introduction

Energy demand is expected to increase by 34% by 2035 as the world’s population grows rapidly and urbanization occurs [1]. The resulting increase in energy consumption will substantially enhance carbon emissions and accelerate climate change, which is a major threat to society. To address climate threats, both the Paris Agreement and the Sustainable Development Goals (SDGs) have proposed long-term greenhouse gas reduction strategies, and many scholars have proposed various initiatives to reduce carbon emissions [2]. Most researchers believe that carbon neutrality refers to a state of net-zero emissions following efforts to reduce or offset carbon emissions [3,4].
Transport has become the second largest source of carbon emissions, accounting for over 24% and continuing to grow rapidly [1]. Enhancing the carbon reduction effect (CRE) in transport infrastructure and understanding the factors influencing this effectiveness have sparked the interest of many researchers. The literature identifies several factors influencing the effectiveness of carbon reduction strategies. These factors fall into three main categories: economic, institutional, and technological. Most studies have focused on energy and industry while placing less emphasis on transport infrastructure; moreover, they have not examined the impact of the social environment and use of digital control technologies (DC) on the effectiveness of carbon reduction strategies [5,6,7]. Khan et al. [8] showed that low-carbon targets can be achieved by improving the capacity and effectiveness of energy management. Therefore, maximizing carbon reductions and improving carbon emission management are important for achieving carbon neutrality targets and promoting the sustainable development of transport infrastructure. Alola et al. [9] noted that low-carbon awareness contributes to sustainable social development. Chen et al. [5] pointed out that DC have a positive effect on the achievement of carbon neutrality targets. However, the contributions of the various influencing factors behind the effectiveness of carbon reductions in transport infrastructure remain unexamined. The literature on the role of the social environment and DC in the carbon reduction process within transport infrastructure is inadequate. Finding answers to these questions thus requires further research.
Many countries have committed to carbon reduction and neutrality in response to the Paris Agreement and to achieve the Sustainable Development Goals (SDGs). With economic development, the transportation sector has become an important source of CO2 emissions. As an important carrier of the transport industry, the transport infrastructure is of great significance in controlling carbon emissions from this sector and achieving carbon neutrality and the SDGs targets. However, most studies have focused on transport vehicles and neglected the transport infrastructure itself. Furthermore, the influence of collusive behavior and digital control on the process of carbon reduction has also not been examined. The novelties of this paper are as follows: (1) the subject of the study is the transport infrastructure, rather than the common vehicle, extending the existing research perspectives; and (2) in the construction of the research model, the consideration of collusive behavior and the impact of digital regulation is incorporated.
This study aims to address the following questions:
Do economic, institutional, and technological factors still play a role in the carbon reduction process in transport infrastructure?
Does collusive behavior mediate the relationships among industrial structure, low-carbon regimes, and the effectiveness of carbon reductions?
Do DC mediate the effects of low-carbon regimes and LCTs on carbon reductions?
In the second part of this study, a literature review is conducted to summarize the relevant studies on carbon emission reduction in transport infrastructure, which provides the basis for model construction; in the third part, the basic hypotheses of the study are put forward on the basis of the analysis of the literature; in the fourth and fifth parts, the constructed model is analyzed and verified by using the PLS-SEM method, and the conclusions are obtained; and in the sixth part, the conclusions are summarized, recommendations are made for the research issues, and the limitations of the study are explained. problems, and the limitations of the study are explained.

2. Literature Review

2.1. Carbon Neutrality Studies

Growing energy demand has impacted global climate change. In response to climate change, many countries worldwide have proposed carbon neutrality targets, and more than 30 have established explicit carbon emission plans [10]. Carbon neutrality is the achievement of net-zero carbon emissions within a specified period [11] (IPCC, 2020). However, many other countries have not yet set carbon neutrality goals. Australia and New Zealand were the first countries to develop and implement carbon-neutral systems (called “zero-carbon emission systems” in New Zealand). Only a few countries, such as Norway and Denmark, refer to their carbon neutrality goals as “climate neutral”, and most countries, including China, refer to them as a “carbon neutrality target” in their policies [12].
Many scholars have studied methods of achieving carbon neutrality. The path to achieve the dual carbon goal comprises the following two steps: first, reduce carbon sources and greenhouse gas emissions, mainly carbon; second, increase carbon sinks through various means and technologies to increase the amount of carbon absorbed [13]. Regarding infrastructure, using energy-efficient materials and technological innovations can reduce carbon emissions, while cutting-edge technology is essential for achieving carbon neutrality goals [14].
Worldwide, countries are gradually turning carbon neutrality targets into national strategies and actively adopting carbon reduction measures to achieve net-zero carbon emissions within a specified period. Therefore, it is of theoretical and practical significance to explore the factors influencing carbon reductions. Only a few scholars have focused their carbon reduction research on the main body of transport infrastructure under the carbon neutrality target. Most have studied the factors influencing carbon reductions from power, transportation, steel, and other industries [15,16,17]. These show that a carbon neutrality target can only be achieved through the combined effect of technological innovation and industrial restructuring [18,19].

2.2. Carbon Emissions from Transport Infrastructure

In the transport industry, transport infrastructure not only plays an important role in supporting and guaranteeing socioeconomic development but is also a major source of carbon emissions [20]. Transport infrastructure generates large amounts of carbon emissions during the construction, operation, and maintenance phases, which increase with the increase in such infrastructure. Further, with continuous economic development, demand for transport infrastructure will grow rapidly. To accommodate this rising demand, controlling carbon emissions from transport infrastructure has become an urgent issue. Therefore, the control and management of carbon emissions from transport infrastructure can help promote overall carbon reduction [21,22]. Carbon emissions from transport include not only emissions from transport infrastructure, which are fixed sources such as roads, ports, airports, etc., but also emissions from transport equipment, which are mobile sources such as cars, trains, and lorries, and the factors influencing carbon emissions from transport infrastructure are more related to the infrastructure itself [23,24]. Therefore, the factors influencing them are also quite different. Carbon reduction in transport infrastructure can also be achieved through the promotion of sustainable modes of transport, with scholars focusing on emissions modeling, travel patterns, shared mobility, and environmental monitoring of roads for sustainable transport [25,26,27] However, most existing studies have focused on measuring carbon emissions and analyzing the factors influencing them in the transportation sector, including exploring the impact of the type of transport, volume of transport, and mode of energy supply (e.g., oil, electricity) on carbon emissions [24]. By contrast, only a few scholars have focused on the factors influencing carbon emissions in transport infrastructure.

2.3. Carbon Emission-Related Digital Control Technologies

Since DC play an important role in the real-time monitoring of carbon emissions in energy production, transportation, and consumption, it can be assumed that they help achieve carbon neutrality. With the continuous advancement of research on this topic, emerging technologies such as big data, digital twins, artificial intelligence, and blockchain technology have been increasingly adopted to mitigate carbon emissions.
In practical applications, DC can regulate carbon emissions by analyzing the capacity for carbon capture [3,28]. On this basis, a carbon emission trading platform can be constructed based on the principles of “decentralization, transparency, security, non-tampering, and traceability” of Internet and blockchain technology [3]. Enterprises with carbon emissions above or below the set quota can buy or sell carbon offsets on the platform [29]. However, the use of DC also generates carbon emissions owing to their high power consumption and high dependence on energy localization [30]. Research largely ignores carbon emissions from digital technology, which lowers the efficiency of carbon reduction processes.

2.4. Impact of Collusive Behavior on Carbon Emissions

Infrastructure construction suffers from a high incidence of collusion. In carbon emission regulation, there is not only government–enterprise collusion but also collusion between enterprises and third-party institutions. The occurrence of these behaviors adversely affects the control of carbon emissions. Collusive behavior is defined in this paper as “two or more entities colluding to violate relevant policies and engage in carbon-intensive behaviors in order to achieve higher revenues for themselves without being penalized.” Research reveals a dynamic relationship between regional collusion and carbon emissions in developing countries: regional collusion first leads to an increase in carbon emissions, which is unfavorable to carbon reductions, and then drives up carbon emissions [31]. Third-party verification agencies collude with companies to falsely report carbon emissions for profit, which seriously hinders the carbon reduction process [32]. However, research largely focuses on preventing the occurrence of collusive behavior and formulating punitive measures for when it occurs, as the occurrence of such behavior results in economic losses and reduced environmental performance that are difficult to recover from even if they are remedied. Therefore, detecting ongoing collusion and combating it remains challenging.
The existing literature on the reduction of carbon emissions in the field of transport has been reviewed, and it can be seen that the majority of research has focused on vehicles, traveling modes, planning, and other related issues. There is, however, a paucity of studies on the impact of emission reduction on transport infrastructure. Furthermore, there is a dearth of research that has sought to design how multiple factors work together in order to achieve emission reduction. This research paper seeks to address these shortcomings.

3. Research Hypotheses

3.1. Impact of the Industrial Structure on Carbon Reductions

The traditional high-carbon industrial structure (IS) and development model are major drivers of the increased carbon emissions in China. In the context of carbon neutrality, traditional generation structures and development models urgently need to be optimized and updated. The term “industrial structure” refers to the composition, interrelationships, proportions, and links among industries in the national economy of a country or region. The IS has a direct impact on the energy consumption and carbon emissions of each industry. By optimizing and adjusting the industrial structure, the overall industrial system can generate economic benefits while reducing environmental pressure, thus achieving the low-carbon transformation of the industrial chain [33]. The increase in the share of the tertiary sector and economic growth play important roles in curbing carbon emission intensity [34]. Therefore, the rationalization of the IS can improve carbon efficiency and reduce carbon emissions, which are the key drivers of carbon reductions [35,36]. Zhou et al. [37] pointed out that industrial restructuring can reduce carbon emissions. However, during industrial restructuring, the IS is influenced by the collusion of economic agents bound by a uniform scale of competition [38]. Therefore, collusive behavior during industrial restructuring is not conducive to the development of a structure toward rationalization [39]. Controlling collusive behavior can thus help drive the IS in a low-carbon direction. Based on the foregoing, we propose the following hypotheses:
Hypothesis 1 (H1): 
The IS affects carbon reductions.
Hypothesis 2 (H2): 
The IS affects collusive behavior.

3.2. Impact of Low-Carbon Technologies on Carbon Reductions

In transport infrastructure, low-carbon technology (LCT) refers to the methods used in the design, construction, operation, and maintenance of transport infrastructure projects that reduce carbon emissions. Increased spending on technology can reduce carbon emissions to a limited extent; however, in today’s climate, breaking the technology bottleneck and developing LCTs are the key factors to reducing carbon emissions from infrastructure [6]. However, if it is not economical to adopt low-carbon technologies, the transition to alternative technologies such as low-carbon fuel substitution technologies will occur, not only to reduce carbon intensity but also to help dispose of waste. In the context of carbon neutrality goals, the implementation of LCTs can enhance infrastructural development, making it necessary to promote the innovation and deployment of low-carbon technologies and replace current high-carbon technologies with these alternatives to reduce energy and carbon intensities [40]. Therefore, adopting LCTs can reduce carbon emission intensity and achieve greater carbon reductions; however, LCTs are affected by policies, expenditure, and application prospects.
DC (e.g., digital management platforms and carbon emission monitoring systems) can be used to control carbon emissions. Such technologies provide safe and reliable technical support for energy production, promote the innovation of low-carbon technologies, and establish a digital operation and control system, which can also monitor and manage greenhouse gases, thus helping reduce carbon emissions [3]. In addition, the application of LCTs in transport infrastructure has been found to reduce the cost of government and improve its efficiency, thus effectively avoiding collusive behavior between local governments and enterprises and prompting enterprises to reduce emissions [41]. Consequently, LCTs, DC, and collusive behavior also play a role in promoting carbon reductions. Based on the foregoing, we propose the following hypotheses:
Hypothesis 3 (H3): 
LCTs affect the effectiveness of carbon reductions.
Hypothesis 4 (H4): 
LCTs affect DC.
Hypothesis 12 (H12): 
LCTs indirectly affect the effectiveness of carbon reductions through collusive behavior.

3.3. Impact of Low-Carbon Regimes on Carbon Reductions

Transport infrastructure continues to have high levels of energy consumption and emissions. Reducing carbon in transport infrastructure requires the joint action of multiple factors. Many researchers have investigated whether current low-carbon policies can help achieve carbon reduction goals and the reasons that hinder the implementation of low-carbon policies. From the government’s perspective, inadequate laws and regulations related to carbon taxes and the insufficient implementation of low-carbon policies are the main factors restricting the development of low-carbon buildings [42]. Limited investment budgets and a lack of knowledge and experience of LCTs are thus important reasons for this inability to implement low-carbon policies [43]. However, most scholars believe that low-carbon policies can reduce the carbon emissions of enterprises.
When enterprises carry out production activities under a low-carbon system, an overly strict carbon trading system weakens enterprises’ carbon reductions. In this regard, to compensate for their reduced profits, in the absence of prior communication, jointly choose the strategy of raising prices to avoid losses. The choice of enterprises to follow low-carbon policy is not constrained only by the carbon reduction target, as the government’s DC significantly mediate the effect of an enterprise’s carbon reductions. DC, which the government uses to review the carbon emissions of enterprises to calculate real carbon emission data as well as avoid collusion between enterprises and third-party verification agencies.
DC provides the basic support for the application of LCTs and the implementation of LCRs. After the development of LCRs, strong implementation is still needed to play a proper role, so it is necessary to provide a guarantee for them through DC, and the development of LCRs will also promote the application of DCs. The existence of DC will suggest the effect of LCTs in the role of CRE.
Based on the aforementioned, the following hypotheses are proposed:
Hypothesis 5 (H5): 
Low-carbon regimes affect carbon reductions.
Hypothesis 6 (H6): 
Low-carbon regimes affect DC.
Hypothesis 7 (H7): 
Low-carbon regimes affect collusive behavior.
Hypothesis 11 (H11): 
Low-carbon regimes indirectly affect carbon reductions through collusive behavior.

3.4. Impact of Digital Control Technologies on Carbon Reductions

As DC continue to develop in the environmental field, their role in achieving carbon neutrality goals is receiving increasing attention. DC are not as dependent on natural resources as traditional production technologies are. Hence, their innovation can help set carbon accounting standards, promote the application of new energy technologies, and develop carbon absorption systems. Therefore, low-carbon systems can function better and solve the difficulties of low-carbon development, thus promoting the achievement of carbon neutrality goals [4,44]. Liu et al. [45] found that DC can not only reduce local carbon emissions but also promote carbon reductions in neighboring cities. Therefore, innovation in DC reduces carbon emissions intensity nationally, and when technological development reaches a certain threshold, the effectiveness of carbon reductions gradually becomes obvious [46,47]. Promoting DC based on artificial intelligence and blockchain can thus promote carbon reductions in transport infrastructure. Based on the foregoing, we propose the following hypotheses:
Hypothesis 8 (H8): 
Low-carbon regimes indirectly affect carbon reductions through DC.
Hypothesis 9 (H9): 
LCTs indirectly affect carbon reductions through DC.
Hypothesis 10 (H10): 
DC affect carbon reductions.

3.5. Impact of Collusive Behavior on Carbon Reductions

The presence of collusive behavior impacts the implementation of government policies and the intensity of management and control, as well as the effectiveness of carbon reductions in infrastructure. Achieving the SDGs and dual carbon goal simultaneously can promote industrial upgradation at the macro level; however, it can damage the economic benefits of enterprises using the original high-carbon development model in the short term. Therefore, companies may be motivated to adopt collusion strategies that hinder the implementation of low-carbon policies. Third-party institutions in carbon emission regulations play an important role in maintaining the global carbon market [48]. Their supervision can help enterprises mitigate carbon emissions, while they can also establish a carbon emission management system and provide data support for the carbon trading market. Early carbon verification is crucial because it affects the allocation of carbon quotas to enterprises.
If an enterprise chooses to collude with a third-party organization, it can obtain more carbon allowances by increasing the falsely reported carbon emissions of industries that have been eliminated, thereby increasing income. Therefore, using DC can control collusion in the processes of carbon emission regulation and reduction. Using LCTs to create an open, transparent, and democratic environment can reduce this degree of collusion [49]. However, some studies have found that enterprises are more likely to collude when third-party verification agencies are too strict. At this time, the government should build multilevel safeguards to reduce collusive behavior [32]. Based on the foregoing, the following hypothesis is proposed:
Hypothesis 13 (H13): 
Collusive behavior affects carbon reductions.

4. Materials and Methods

4.1. Model Design

A research model was constructed using the factors affecting the reduction of carbon emissions from transport infrastructure based on the foregoing literature review and hypotheses. This is illustrated in Figure 1.

4.2. Questionnaire Design

Through group discussions, expert interviews, and on-site and online collections, this study designed a questionnaire on the effectiveness of carbon reductions. It was pre-tested in Changsha, and the questionnaire was adjusted and improved based on the results to reflect respondents’ opinions more comprehensively and correctly. The questionnaire included items based on the following six latent variables: the IS, LCTs, low-carbon regimes, DC, collusive behavior, and CRE; the survey language was Chinese (Table 1). All the items were obtained from previous studies to ensure the validity of the questionnaire. A five-point Likert scale (5 = strongly disagree, 4 = disagree, 3 = no opinion, 2 = agree, and 1 = strongly agree) was used for responses. The gender, occupation, and years of experience of respondents were also recorded. Participation in the questionnaire was voluntary, and the purpose of the study was explained to all participants. Participants’ information and responses were kept confidential during the survey to reduce the threat of common method bias [50].

4.3. Demographic Information

The research team conducted the research from October 2021 to July 2022 with government agencies and companies related to transport infrastructure. An effective response rate of 93.9% was achieved based on 430 questionnaires returned.
The occupational distribution of the respondents was as expected, comprising 6.19% working in the government, 53.46% in transport infrastructure (investment, construction, and operation), 18.32% in research, 15.84% in carbon emissions supervision and management, and 6.19% other. Of the respondents, 59.4% had more than five years of work experience. SPSS AU was used to collate the data. Table 2 presents the characteristics of the surveyed population.

4.4. Methodological Options

Structural equation modeling (SEM) is a second-generation statistical modeling method that allows the quantitative analysis of unobservable variables and can be used to assess the initial assumptions of causal models [65]. Common SEM methods include covariance-based SEM (CB-SEM) and partial least squares SEM (PLS-SEM) [66,67]. Both methods have their applications and limitations. CB-SEM is mainly used for testing and confirming hypotheses; it emphasizes fit indices and is suitable for analyzing large samples. PLS-SEM is usually used to analyze the correlations between relevant hypotheses and theories and is more concerned with predicting causality and assessing quality. It is used when (1) a theoretical framework is being tested from a predictive perspective, (2) a complex problem is being addressed through the exploration of established theories, and (3) minority groups limit the number of participants. According to the above, PLS-SEM can be used to test the validity of the model and to analyze the causal relationship between variables that are difficult to observe, which is in line with this paper’s analysis of the relationship between LCT, DC, LCR, CB, IS, and CRE; at the same time, according to the existing research, PLS-SEM is more reliable in the case of multi-structures, which meets the requirements of the model setting in this paper. Considering the characteristics of the subjects analyzed, purpose of the study, and quality of the data, PLS-SEM was chosen as the research method for this study.

5. Result

5.1. Methods

To enhance the model’s ability to handle complexity and establish predictive validity, this study used the bootstrapping technique of PLS-SEM to test and examine the measurement and structural models [67,68,69]. This study was analyzed using SmartPLS 3 software.
For the measurement model, the focus was on analyzing its reliability. We assessed the Cronbach’s alpha, a method of measuring the reliability of a scale or test that overcomes the shortcomings of the partial halving method and is the most commonly used method of reliability analysis in scientific research, composite reliability (CR; the reliability of a composite score, a new variable consisting of the sum of more than one variable), t-statistics, factor loadings (used to assess the reliability and credibility of the strategy model, indicating the strength of the relationship between the measure and the latent variable), and covariance structures of the model to examine its reliability. In an exploratory analysis, when Cronbach’s α > 0.7 and CR > 0.7, the results are reliable; hence, our measurement model had good reliability [70]. Moreover, all the factor loadings in this study were above 0.5, suggesting that the variables were explained reasonably well. The model also had a variance inflation factor (VIF) below 5 [71], meeting the requirements in this regard. Finally, we found that the observed variables were linearly related to the latent variables, which makes the latent variables easier to understand.
The goodness-of-fit concept commonly used in CB-SEM does not have a universal value for PLS-SEM models. Hence, the standardized root mean square residual (SRMR), heterotrait-monotrait (HTMT) ratio, and normed fit index (NFI) were selected for this study [72] (see Table 3 and Table 4).
The closer the NFI is to 1, the better is the fit of the model. However, a major disadvantage is that it does not penalize for model complexity. The higher the number of parameters in the model, the larger (i.e., better) the NFI result. A threshold of above 0.9 has been selected by some scholars as a good NFI, with >0.8 or >0.85 an acceptable goodness-of-fit criterion in some studies; other scholars also consider it acceptable when the NFI is between 0.7 and 0.9. In this study, the NFI was 0.875. Considering the study and survey characteristics, this indicator was thus considered acceptable.
However, Hair et al. [79] pointed out that using the NFI for analyzing models in PLS-SEM is not recommended. Therefore, we also used additional metrics to determine the reliability of the model (i.e., SRMR and HTMT). As shown in Table 3, the SRMR and HTMT values of the model met the requirements, suggesting that the model had sufficient discriminant validity (Table 4). After the comparison of several indicators, we concluded that the proposed model was suitable for use in the subsequent analysis.
We subsequently calculated the factor loadings and R2 values of the model (Figure 2) to analyze its reliability and construct validity. We found that the reliability, convergent validity, discriminant validity, and significance level of the model were acceptable. According to existing studies, the conclusions of a model are reliable when Cronbach’s α and CR are above 0.7 in PLS-SEM [70]. In the model, all the items had VIFs below 5, which met the requirements without further adjustments to the model (Table 5). We can measure the convergent and discriminant validity of a PLS-SEM model using the AVE, which should be above 0.5; further, the square root of the AVE should be above the correlation coefficients of the other latent variables (Table 6 and Table 7). Hence, it was concluded that the measured variables explained the latent variables well and that the model construction was reasonable.
Discriminant validity is an important method of assessing the validity of a PLS-SEM model to determine whether there is a difference between constructs in a research model. The Fornell-Larcker criterion is an important method of assessing discriminant validity, where the AVE for each construct should be greater than the correlation coefficient between that construct and the rest of the constructs in the model. It ensures that there is no overlap between constructs. According to Table 7, it can be found that the indicators in the constructed model are independent from each other.
R2 is usually used to evaluate the degree to which the independent variables of a model explain the variance of the dependent variable, while Q2 is used to evaluate the predictive relevance of the model and analyze the relationship between the power of the full measurement model and the quality of the full structural model by calculating the goodness-of-fit [5,80]. The model in this study had a good fit (0.582); values over 0.26 indicate that a model has a high goodness-of-fit, which suggests good applicability to social surveys. The R2 values for collusive behavior, CRE, and DC were 0.512, 0.566, and 0.517, respectively, indicating that the model explained a moderate degree of variance. Further, the Q2 was above zero, indicating that the model had good predictive relevance and that the structural model used in this study was robust (see Table 8).

5.2. Analysis of the Results

When using PLS-SEM, we typically test hypotheses by analyzing the t- and p-values of the paths [80]. In the analysis of the path coefficients, statistical significance was assessed using the t-value, where t > 1.96 reached a significance level of p < 0.05, t > 2.58 reached a significance level of p < 0.01, and t > 3.29 reached a significance level of p < 0.001. As shown in Figure 3, Table 9 and Table 10, the coefficients of the paths between the IS and collusive behavior as well as between the IS and carbon reductions were 0.074 (t = 1.498 < 1.96) and 0.317 (t = 6.619 > 1.96), respectively, indicating that Hypothesis 1 was not supported and Hypothesis 2 was supported. However, Table 10 shows that the coefficient of IS → CB → CRE was 0.05 (t = 2.74 > 1.96), supporting Hypothesis 12. The results indicate that the IS does not directly affect the effectiveness of carbon reductions but can promote it through its effect on the social environment. This is consistent with the findings of Chen et al. [5] that the social environment positively affects the control of carbon emissions. The coefficients of the paths from low-carbon technologies to the effectiveness of carbon reductions and DC were 0.200 (t = 3.900 > 1.96) and 0.386 (t = 8.905 > 1.96), respectively, supporting Hypotheses 3 and 4. LCTs affect carbon reductions and DC. When low-carbon technologies are applied, DC and the effectiveness of carbon reductions improve to a greater extent. The coefficients of the paths from low-carbon systems to the effectiveness of carbon reductions, digital control technologies, and collusive behavior were 0.318 (t = 5.525 < 1.96), 0.394 (t = 8.555 > 1.96), and 0.457 (t = 9.997 > 1.96), respectively. Hence, low-carbon systems can promote the social environment and progress in DC, enhancing the effectiveness of carbon reductions. Hypotheses 6 and 7 were supported. The effect of DC on carbon reductions was 0.119 (t = 2.340 > 1.96), supporting Hypothesis 10. Thus, the application of low-carbon technologies directly affects carbon reductions. The coefficient of the path between the social environment and the effectiveness of carbon reductions was 0.157 (t = 3.106 > 1.96); hence, Hypothesis 13 was supported.
Next, DC and collusive behavior played mediating roles in the relationships among LCTs, low-carbon regimes, and CRE. This implies that with better DC and less collusive behavior, LCTs and low-carbon regimes play greater roles in reducing carbon emissions.
While the use of low-carbon technologies will contribute to a reduction in carbon emissions, the initial research and development phase will entail additional costs for enterprises compared to the use of high-carbon technologies. Furthermore, due to the lack of sufficient validation, there are reliability and market acceptance issues with low-carbon technologies at this early stage of their application, which will hinder their adoption. The use of digital control technology in regulation allows for a more comprehensive and rigorous control process. It reduces the likelihood of companies taking chances and influencing their choice of low-carbon behavior; it reduces long-term costs for the regulator.

6. Discussion and Limitations

6.1. Discussion

This study verifies that LCT, DC, and collusive behavior have positive direct and indirect impacts on CRE; DC can enhance the effectiveness of carbon emission reduction by effectively enhancing the effectiveness of the measures, whereas CB will firstly exacerbate carbon emissions and inhibit the effectiveness of carbon emission reduction to a certain extent, consistent with the literature [57,58]. First, low-carbon and DC directly affect the technologies available to transport infrastructure companies in the carbon reduction process and the cost of technology use, which can significantly influence the possibility and cost of using carbon reduction technologies. Simultaneously, the application of DC increases the cost of non-compliance for these enterprises and enhances the possibility that their operations are standardized. When most enterprises choose to operate at a low-carbon level, high-carbon enterprises will also transform, thus improving the overall effectiveness of carbon reductions across society. Meanwhile, the IS does not directly affect CRE. Transport infrastructure companies at the end of the chain lack a comprehensive and correct understanding of the IS and policies of the whole industry. Therefore, the IS can only influence the decisions of transport infrastructure companies through other factors that can be perceived by them.
DC are shown to mediate the process by which low-carbon regimes and LCTs influence CRE. This mediating role of DC was also found by previous studies [61,62]. First, the implementation of low-carbon regimes requires the involvement of controls, and DC can improve the efficiency of controls and process compliance. This increases the likelihood of detecting high carbon emissions by transport infrastructure companies and raises the cost of non-compliance, all of which can improve CRE. Collusive behavior also plays a mediating role in the influence of low-carbon regimes and the IS on CRE. This demonstrates that external IS and the introduction of low-carbon regimes can affect the low-carbon behavior of transport infrastructure firms by influencing the social environment. This finding has both theoretical and practical implications. In particular, it deepens previous research on the causes of enterprises’ low-carbon behavior and explains that the social environment and DC play a role in enhancing CRE, which can provide a basis for future policy formulation.
These findings provide a strong theoretical basis and useful perspectives for the improvement of environmental quality and public welfare in China as well as the mitigation of climate change. First, the IS, LCTs, low-carbon systems, and low-carbon regimes all affect CRE directly or indirectly. On the one hand, this means that enterprises can control carbon emissions within the government’s limit and meet the dual carbon goal. On the other hand, enterprises can reduce their expenditure on carbon taxes. Further, the government can gradually achieve the goals to which China committed at the UN General Assembly, reflecting the country’s responsibility as a great power. For the public, reducing collusive behavior can improve social welfare, while effective carbon reductions can improve the quality of people’s living environment.
Second, LCTs, DC, and collusive behavior have a direct impact on CRE. The government and relevant departments, with the support of national development strategy, should continue to promote the development of low-carbon and DC in transport infrastructure, thus improving the effectiveness of carbon reductions in society and making China’s economic development sustainable and healthy. Moreover, collusive behavior is shown to play a mediating role in the effect of low-carbon regimes and the IS on CRE. In areas with low carbon reductions, local governments and related departments should strictly control the relocation of highly polluting enterprises to these areas; in areas with high carbon reductions, it is necessary to develop and introduce LCTs that match the IS as well as gradually improve the effectiveness of carbon reductions through continuous exchanges among cooperative enterprises.

6.2. Limitations

This study validated the direct and indirect factors influencing CRE; however, it has certain limitations. First, as it is only a preliminary study of the factors influencing CRE in transport infrastructure, the measures and influencing factors may be less than perfect. Future research could aim to refine the items of the influencing factors. Second, while this study examines the mediating roles of DC and collusive behavior on CRE, other mediating variables could impact CRE, which future research could pursue. Third, this study is based in China; future research could expand the scope of the research to verify the theory proposed in this study.

7. Conclusions

In this study, PLS-SEM was used to analyze the factors influencing CRE in transport infrastructure and explore the mediating roles of these influencing factors. Based on the analysis, two main conclusions can be drawn. First, low-carbon technologies, DC, and collusive behavior have direct and positive influences on CRE. The coefficients of the paths from low-carbon technologies to the effectiveness of carbon reductions and DC were 0.200 (t = 3.900 > 1.96) and 0.386 (t = 8.905 > 1.96), respectively. Hence, LCT affect carbon reductions and DC. The coefficients of the paths from low-carbon systems to CRE, DC, and collusive behavior were 0.318 (t = 5.525 < 1.96), 0.394 (t = 8.555 > 1.96), and 0.457 (t = 9.997 > 1.96), respectively. Hence, low-carbon systems can promote the social environment and innovation in DC, enhancing the effectiveness of carbon reductions.
Second, DC were shown to play a mediating role in the influence of low-carbon regimes and LCTs on CRE, while collusive behavior also mediated the influence of low-carbon regimes and the IS on CRE. The coefficients of the paths from the IS to collusive behavior and carbon reductions were 0.074 (t = 1.498 < 1.96) and 0.317 (t = 6.619 > 1.96), respectively. The coefficient of IS → CB → CRE was 0.05 (t = 2.74 > 1.96). Finally, the IS does not directly influence the CRE but does promote it through its effect on the social environment.

Author Contributions

Conceptualization, Y.C. and D.W.; methodology, D.W.; software, D.W.; validation, C.M.; formal analysis, D.W.; investigation, Y.C.; resources, A.W.; data curation, M.H. and Q.L.; writing—original draft preparation, D.W.; writing—review and editing, Y.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data are not publicly available, as they contain information that could compromise the privacy of the research participants.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. IEA. World Energy Outlook; IEA: Paris, France, 2015. [Google Scholar]
  2. Blau. The Paris Agreement; Palgrave MacMillan: London, UK, 2017. [Google Scholar]
  3. Chen, X.; Hu, D.; Cao, W.; Liang, W.; Xu, X.; Tang, X.; Wang, Y. Path of digital technology promoting realization of carbon neutrality goal in China’s energy industry. Bull. Chin. Acad. Sci. 2021, 36, 1019–1029. [Google Scholar]
  4. Zhao, X.; Ma, X.; Chen, B.; Shang, Y.; Song, M. Challenges toward carbon neutrality in China: Strategies and countermeasures. Resour. Conserv. Recycl. 2022, 176, 105959. [Google Scholar] [CrossRef]
  5. Chen, Y.; Wang, D.; Zhu, W.; Hou, Y.; Liu, D.; Ma, C.; Li, T.; Yuan, Y. Effective conditions for achieving carbon unlocking targets for transport infrastructure development: Joint analysis based on PLS-SEM and NCA. Int. J. Environ. Res. Public Health 2023, 20, 1170. [Google Scholar] [CrossRef]
  6. Seto, K.C.; Davis, S.J.; Mitchell, R.B.; Stokes, E.C.; Unruh, G.; Ürge-Vorsatz, D. Carbon lock-in: Types, causes, and policy implications. Annu. Rev. Environ. Resour. 2016, 41, 425–452. [Google Scholar] [CrossRef]
  7. Zaid, S.M.; Myeda, N.E.; Mahyuddin, N.; Sulaiman, R. Malaysia’s rising GHG emissions and carbon ‘lock-in’ risk: A review of Malaysian building sector legislation and policy. J. Surv. Constr. Prop. 2015, 6, 1–13. [Google Scholar]
  8. Khan, Z.; Muhsin, A.; Liu, J.; Muhammad, S.; Yang, S. Consumption-based carbon emissions and trade nexus: Evidence from nine oil exporting countries. Energy Econ. 2020, 89, 104806. [Google Scholar] [CrossRef]
  9. Alola, A.A.; Okere, K.I.; Muoneke, O.B.; Dike, G.C. Do bureaucratic policy and socioeconomic factors moderate energy utilization effect of net zero target in the EU? J. Environ. Manag. 2022, 317, 115386. [Google Scholar] [CrossRef]
  10. van Soest, H.L.; den Elzen, M.G.J.; van Vuuren, D.P. Net-zero emission targets for major emitting countries consistent with the Paris Agreement. Nat. Commun. 2021, 12, 2140. [Google Scholar] [CrossRef]
  11. IPCC. Global Warming of 1.5 °C. 2020. Available online: https://www.ipcc.ch/sr15/chapter/spm/ (accessed on 24 September 2024).
  12. Gerretsen, M.D.A.I. Which Countries Have a Net Zero Carbon Goal? Climate. 2021. Available online: https://www.climatechangenews.com/2019/06/14/countries-net-zero-climate-goal/ (accessed on 6 September 2022).
  13. Zeng, Y.; Wang, X.; Tang, H.; Liao, J.; Tian, M. A review of scientific connotation, realization path and challenges of CO2 emission peak and neutrality. Mod. Chem. Ind. 2022, 6, 1–10. [Google Scholar]
  14. Chapman, A.; Ertekin, E.; Kubota, M.; Nagao, A.; Bertsch, K.; Macadre, A.; Tsuchiyama, T.; Masamura, T.; Takaki, S.; Komoda, R.; et al. Achieving a carbon neutral future through advanced functional materials and technologies. Bull. Chem. Soc. Jpn. 2022, 95, 73–103. [Google Scholar] [CrossRef]
  15. Cao, Y.; Wang, Y. Smart carbon emission scheduling for electric vehicles via reinforcement learning under carbon peak target. Sustainability 2022, 14, 12608. [Google Scholar] [CrossRef]
  16. Carley, S. Historical Analysis of U.S. electricity markets: Reassessing Carbon Lock-in. Energy Policy 2011, 39, 720–732. [Google Scholar] [CrossRef]
  17. Zoll, M. Path-Dependencies of Carbon Lock-In Shaping Coal Phase-Out in Poland’s Electricity Sector: A Herculean Task of Decarbonization? Springer International Publishing: Cham, Switzerland, 2013; pp. 341–368. [Google Scholar]
  18. Wang, C.; Zhang, Y. Implementation pathway and policy system of carbon neutrality vision. Chin. J. Environ. Manag. 2020, 12, 58–64. [Google Scholar] [CrossRef] [PubMed]
  19. Dong, B.; Ma, X.; Zhang, Z.; Zhang, H.; Chen, R.; Song, Y.; Shen, M.; Xiang, R. Carbon emissions, the industrial structure and economic growth: Evidence from heterogeneous industries in China. Environ. Pollut. 2020, 262, 114322. [Google Scholar] [CrossRef] [PubMed]
  20. Dominković, D.F.; Bačeković, I.; Pedersen, A.S.; Krajačić, G. The future of transportation in sustainable energy systems: Opportunities and barriers in a clean energy transition. Renew. Sustain. Energy Rev. 2018, 82, 1823–1838. [Google Scholar] [CrossRef]
  21. Li, H.; Luo, N. Will improvements in transportation infrastructure help reduce urban carbon emissions? Motor vehicles as transmission channels. Environ. Sci. Pollut. Res. 2022, 29, 38175–38185. [Google Scholar] [CrossRef]
  22. Xie, R.; Fang, J.; Liu, C. The effects of transportation infrastructure on urban carbon emissions. Appl. Energy 2017, 196, 199–207. [Google Scholar] [CrossRef]
  23. Awaworyi Churchill, S.; Inekwe, J.; Ivanovski, K.; Smyth, R. Transport infrastructure and CO2 emissions in the OECD over the Long Run. Transp. Res. Part D Transp. Environ. 2021, 95, 102857. [Google Scholar] [CrossRef]
  24. Wang, Y.; Zhou, Y.; Zhu, L.; Zhang, F.; Zhang, Y. Influencing factors and decoupling elasticity of China’s transportation carbon emissions. Energies 2018, 11, 1157. [Google Scholar] [CrossRef]
  25. Hsieh, F.-S. A comparison of three ridersharing cost saving allocation schemes based on the number of acceptable shared rides. Energies 2021, 14, 6931. [Google Scholar] [CrossRef]
  26. Macioszek, E.; Granà, A.; Fernandes, P.; Coelho, M.C. New Perspectives and Challenges in Traffic and Transportation Engineering Supporting Energy Saving in Smart Cities—A Multidisciplinary Approach to a Global Problem. Energies 2022, 15, 4191. [Google Scholar] [CrossRef]
  27. Vasiutina, H.; Szarata, A.; Rybicki, S. Evaluating the Environmental Impact of Using Cargo Bikes in Citie: A Comprehensive Review of Existing Approaches. Energies 2021, 14, 6462. [Google Scholar] [CrossRef]
  28. Cao, L. Application of artificial intelligence on the CO2 capture: A review. J. Therm. Anal. Calorim. 2021, 145, 1751–1768. [Google Scholar] [CrossRef]
  29. He, Y.; Song, W. Analysis of the impact of carbon trading policies on carbon emission and carbon emission efficiency. Sustainability 2022, 14, 10216. [Google Scholar] [CrossRef]
  30. Koot, M.; Wijnhoven, F. Usage impact on data center electricity needs: A system dynamic forecasting model. Appl. Energy 2021, 291, 116798. [Google Scholar] [CrossRef]
  31. Akhbari, R.; Nejati, M. The effect of corruption on carbon emissions in developed and developing countries: Empirical investigation of a claim. Heliyon 2019, 5, e02516. [Google Scholar] [CrossRef]
  32. Wu, X.; Liu, P.; Yang, L.; Shi, X. Analysis of the evolution of carbon emission supervision of logistics enterprises based on third-party verification. Stat. Decis. 2022, 38, 184–188. [Google Scholar]
  33. Chen, M.; Gao, H.; Tian, J.; Chen, L. Green transformation of Jiaxing City’s industrial structure driven by synergizing the reduction of pollution and carbon emissions. Environ. Sci. 2024, 1–17. [Google Scholar] [CrossRef]
  34. Zhang, Y.; Liu, Z.; Zhang, H.; Tan, T. The impact of economic growth, industrial structure and urbanization on carbon emission intensity in China Nat. Nat. Hazards 2014, 73, 579–595. [Google Scholar] [CrossRef]
  35. Li, W.; Wang, W.; Wang, Y.; Qin, Y. Industrial structure, technological progress and CO2 emissions in China: Analysis based on the STIRPAT framework. Nat. Hazards 2017, 88, 1545–1564. [Google Scholar] [CrossRef]
  36. Zhang, S.; Li, H.; Zhang, Q.; Tian, X.; Shi, F. Uncovering the impacts of industrial transformation on low-carbon development in the Yangtze River Delta. Resour. Conserv. Recycl. 2019, 150, 104442. [Google Scholar] [CrossRef]
  37. Zhou, X.; Zhang, J.; Li, J. Industrial structural transformation and carbon dioxide emissions in China. Energy Policy 2013, 57, 43–51. [Google Scholar] [CrossRef]
  38. Dijkstra, P.T.; Haan, M.A.; Mulder, M. Industry structure and collusion with uniform yardstick competition: Theory and experiments. Int. J. Ind. Organ. 2017, 50, 1–33. [Google Scholar] [CrossRef]
  39. Yang, L. Corruption, marketization and rationalization of industrial structure: An analysis based on dynamic spatial panel model. Areal Res. Dev. 2018, 37, 7–10. [Google Scholar]
  40. Niu, H.; Liu, Z. Measurement on carbon lock-in of China based on RAGA-PP model. Carbon Manag. 2021, 12, 451–463. [Google Scholar] [CrossRef]
  41. Fan, R.; Xing, Y. Analysis on game among agents in carbon reduction process of energy-intensive enterprise considering collusion and reputation. J. Technol. Econ. 2016, 35, 81–87. [Google Scholar]
  42. Du, Q.; Yan, Y.; Huang, Y.; Hao, C.; Wu, J. Evolutionary games of low-carbon behaviors of construction stakeholders under carbon taxes. Int. J. Environ. Res. Public Health 2021, 18, 508. [Google Scholar] [CrossRef]
  43. Mustaffa, N.K.; Abdul Kudus, S.; Abdul Aziz, M.F.H.; Anak Joseph, V.R. Strategies and way forward of low carbon construction in Malaysia. Build. Res. Inf. 2022, 50, 628–645. [Google Scholar] [CrossRef]
  44. Hu, Y.; Jin, S. Digital technology helps achieve the goal of ‘double carbon’: Theoretical mechanism and practical path. Financ. Account. Mon. 2022, 6, 111–118. [Google Scholar]
  45. Liu, J.; Yu, Q.; Chen, Y.; Liu, J. The impact of digital technology development on carbon emissions: A spatial effect analysis for China. Resour. Conserv. Recycl. 2022, 185, 106445. [Google Scholar] [CrossRef]
  46. Wang, L.; Chen, Y.; Ramsey, T.S.; Hewings, G.J.D. Will researching digital technology really empower green development? Technol. Soc. 2021, 66, 101638. [Google Scholar] [CrossRef]
  47. Xue, F.; Liu, J.; Fu, Y. The Effect of Artificial Intelligence Technology on Carbon Emissions. Sci. Technol. Prog. Policy 2022, 39, 1–9. [Google Scholar]
  48. Deng, Z.; Gao, T.; Pang, R.; Yang, C. Enterprise passive collusion: Welfare effect analysis of environmental regulation under the goals of carbon peaking and carbon neutrality. China Ind. Econ. 2022, 7, 122–140. [Google Scholar] [CrossRef]
  49. Rodríguez-Martínez, G.-S.; García-Sánchez, G.-V.; Vicente-Galindo, I.M.; Galindo-Villardón, P. Exploring relationships between environmental performance, E-government and corruption: A multivariate perspective. Sustainability 2019, 11, 6497. [Google Scholar] [CrossRef]
  50. Podsakoff, P.M.; Mackenzie, S.B.; Podsakoff, N.P. Sources of method bias in social science research and recommendations on how to control it. Annu. Rev. Psychol. 2012, 63, 539–569. [Google Scholar] [CrossRef]
  51. Jiang, H.; Yu, Z.; Zhang, Y.; Yu, Y.; Huang, W. The temporal and spatial pattern evolution of provincial industrial carbon intensity under the Carbon Neutral Target: Evidence from China. Environ. Sci. Pollut. Res. 2023, 30, 61134–61144. [Google Scholar] [CrossRef]
  52. Nurgazina, Z.; Ullah, A.; Ali, U.; Koondhar, M.A.; Lu, Q. The impact of economic growth, energy consumption, trade openness, and financial development on carbon emissions: Empirical evidence from Malaysia. Environ. Sci. Pollut. Res. 2021, 28, 60195–60208. [Google Scholar] [CrossRef] [PubMed]
  53. Lin, B.; Wang, C. Does industrial relocation affect regional carbon intensity? Evidence from China’s secondary industry. Energy Policy 2023, 173, 113339. [Google Scholar] [CrossRef]
  54. Wang, Z.; Zhang, B.; Liu, T. Empirical analysis on the factors influencing national and regional carbon intensity in China. Renew. Sustain. Energy Rev. 2016, 55, 34–42. [Google Scholar] [CrossRef]
  55. Cary, M. Molecules of inefficiency: How tariffs impact carbon intensities, carbon dioxide emissions, and the environment. Sci. Total Environ. 2020, 713, 136531. [Google Scholar] [CrossRef]
  56. Gao, P.; Wang, Y.; Zou, Y.; Su, X.; Che, X.; Yang, X. Green technology innovation and Carbon Emissions Nexus in China: Does industrial structure upgrading matter? Front. Psychol. 2022, 13, 951172. [Google Scholar] [CrossRef] [PubMed]
  57. Xin, L.; Sun, H.; Xia, X.; Wang, H.; Xiao, H.; Yan, X. How does renewable energy technology innovation affect manufacturing carbon intensity in China? Environ. Sci. Pollut. Res. 2022, 29, 59784–59801. [Google Scholar] [CrossRef] [PubMed]
  58. Wei, W.-X.; Yang, F. The impact of technological progress on carbon dioxide emissions in China. Stat. Res. 2010, 7, 36–44. [Google Scholar] [CrossRef]
  59. Jin, W. Path dependence, self-fulfilling expectations, and carbon lock-in. Resour. Energy Econ. 2021, 66, 101263. [Google Scholar] [CrossRef]
  60. Zhang, X.; He, S.; Ma, L. Local environmental fiscal expenditures, industrial structure upgrading, and carbon emission intensity. Front. Environ. Sci. 2024, 12, 1369056. [Google Scholar] [CrossRef]
  61. Zeng, J.; Yang, M. Digital technology and carbon emissions: Evidence from China. J. Clean. Prod. 2023, 430, 139765. [Google Scholar] [CrossRef]
  62. Chen, J.; Chen, Y.; Liang, X.; Yu, L.; Sun, J.; Luo, X.; Zhang, M.; Wu, Z. The mitigating effect of new digital technology on carbon emissions: Evidence from China. Environ. Sci. Pollut. Res. 2024, 31, 9433–9444. [Google Scholar] [CrossRef]
  63. Yang, Y.; Yang, X.; Tang, D. The dynamic relationship between regional corruption and carbon emissions in China. Clean Technol. Environ. Policy 2020, 25, 223–236. [Google Scholar] [CrossRef]
  64. Gao, W.; Wang, Y.; Wang, F.; Mbanyele, W. Do Environmental Courts break collusion in environmental governance? evidence from Corporate Green Innovation in China. Q. Rev. Econ. Financ. 2024, 94, 133–149. [Google Scholar] [CrossRef]
  65. Koran, J.; Hancock, G.R. Using fixed thresholds with grouped data in structural equation modeling. Struct. Equ. Model. Multidiscip. J. 2010, 17, 590–604. [Google Scholar] [CrossRef]
  66. Fang, W.; Ma, C.; Lei, Z. Research on sustainable development of transport infrastructure based on corporate culture and low-carbon perspective. J. Environ. Public Health 2022, 2022, 4629422. [Google Scholar] [CrossRef]
  67. Rigdon, E.E.; Sarstedt, M.; Ringle, C.M. On comparing results from CB-SEM and PLS-SEM: Five perspectives and five recommendations. Mark. ZFP 2017, 39, 4–16. [Google Scholar] [CrossRef]
  68. Hair, J.F.; Howard, M.C.; Nitzl, C. Assessing measurement model quality in PLS-SEM using confirmatory composite analysis. J. Bus. Res. 2020, 109, 101–110. [Google Scholar] [CrossRef]
  69. Rigdon, E.E. Choosing PLS path modeling as analytical method in European management research: A realist perspective. Eur. Manag. J. 2016, 34, 598–605. [Google Scholar] [CrossRef]
  70. Hair, J.F.; Risher, J.J.; Sarstedt, M.; Ringle, C.M. When to use and how to report the results of PLS-SEM. Eur. Bus. Rev. 2019, 31, 2–24. [Google Scholar] [CrossRef]
  71. Becker, J.; Rai, A.; Ringle, C.M.; Völckner, F. Discovering unobserved heterogeneity in structural equation models to avert validity threats. MIS Q. 2013, 37, 665–694. [Google Scholar] [CrossRef]
  72. Shmueli, G.; Sarstedt, M.; Hair, J.F.; Cheah, J.; Ting, H.; Vaithilingam, S.; Ringle, C.M. Predictive model assessment in PLS-SEM: Guidelines for using PLSpredict. Eur. J. Mark. 2019, 53, 2322–2347. [Google Scholar] [CrossRef]
  73. Bentler, P.M.; Bonett, D.G. Significance tests and goodness of fit in the analysis of covariance structures. Psychol. Bull. 1980, 88, 588–606. [Google Scholar] [CrossRef]
  74. Zhai, J.; Wu, C.; Zhai, C. Efficiency, legitimacy and social network: The logical analysis of rural farmland transfer. J. Yunnan Agric. Univ. (Soc. Sci.) 2020, 14, 82–91. [Google Scholar]
  75. Rong, T. AMOS and Research Methods; Chongqing University Press: Chongqing, China, 2009. [Google Scholar]
  76. Wu, M. Structural Equation Modeling: Operation and Application of AMOS; Chongqing University Press: Chongqing, China, 2009. [Google Scholar]
  77. Wu, L.; Zhao, D.; Wang, X.; Xu, L. Study on manufacturers’ decision-making behavior on carbon-labeled food. China Soft Sci. 2011, 6, 87–99. [Google Scholar] [CrossRef]
  78. Xiong, S. A Study on the Determinants of Chinese Intentions to Use Mobile SNS—A Case Study on Weixin. Bachelor’s Thesis, Beijing University of Posts and Telecommunications, Beijing, China, 2013. [Google Scholar]
  79. Hair, J.F.; Hult, G.T.M.; Ringle, C.M.; Sarstedt, M. A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM); Sage: Los Angeles, CA, USA, 2017. [Google Scholar]
  80. Tenenhaus, M.; Vinzi, V.E.; Chatelin, Y.; Lauro, C. PLS path modeling. Comput. Stat. Data Anal. 2005, 48, 159–205. [Google Scholar] [CrossRef]
Sustainability 16 08390 g001
Figure 1. Theoretical model.
Figure 1. Theoretical model.
Sustainability 16 08390 g001
Sustainability 16 08390 g002
Figure 2. Measurement model with the R2 values.
Figure 2. Measurement model with the R2 values.
Sustainability 16 08390 g002
Sustainability 16 08390 g003
Figure 3. Structural model.
Figure 3. Structural model.
Sustainability 16 08390 g003
Table 1. Latent and observed variables.
Table 1. Latent and observed variables.
Latent VariableObserved VariableReference(s)
ISThe stage at which the industry is developingJiang et al. [51]; Nurgazina et al. [52]; Lin et al. [53]; Wang et al. [54]
R&D investment in low-carbon technologies
Percentage of high-carbon industrial development patterns
LCTCarbon reduction capacity of alternative technologiesCary [55]; Gao et al. [56]; Xin et al. [57]; Wei et al. [58]
Prospects for the application of low-carbon technologies
Level of deployment of low-carbon technologies
Difficulty of trading low-carbon technologies
LCRExpenditure on low-carbon subsidiesJin [59]; Chen, Y. et al. [5]; Zhang et al. [60]
High-carbon development model inertia
DCApplication of digital control technologiesLiu et al. [45]; Zeng et al. [61]; Chen et al. [62]
Construction and operation of digital control technologies
Transformation of the digital management model
CBGovernment–business collusion and policy implicationsYang et al. [63]; Gao et al. [64];
Regional collaborative behavior
Note: Collusion behavior, CB; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 2. Characteristics of the surveyed population.
Table 2. Characteristics of the surveyed population.
VariableTypeValueProportion
OccupationGovernment256.19%
Transport infrastructure investment4811.88%
Transport infrastructure construction7819.31%
Transport infrastructure operations9022.28%
Research7418.32%
Carbon emission supervisors and managers6415.84%
Other256.19%
Years of working<54410.89%
5–1012029.7%
10–1516440.59%
>157618.81%
Table 3. Goodness-of-fit.
Table 3. Goodness-of-fit.
IndicatorValueCriterionReference(s)
SRMR0.054<0.08Bentler and Bonett [73]; Shmueli et al. [72]
NFI0.875>0.85Junliang et al. [74]; Rong [75]; Wu [76]; Wu et al. [77]; Xiong [78]
Note: standardized root mean square residual, SRMR; normed fit index, NFI.
Table 4. HTMT ratio.
Table 4. HTMT ratio.
ISLCRLCTCBDCCRE
IS
LCR0.808
LCT0.7590.803
CB0.7200.7780.755
DC0.7100.7620.7420.724
CRE0.6450.7520.6960.6710.648
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 5. Model items.
Table 5. Model items.
Latent VariableCronbach’s αComposite ReliabilitytpVIFItem
IS0.8760.89719.725<0.0012.323IS1
21.026<0.0012.312IS2
23.618<0.0013.375IS3
LCR0.8540.87018.454<0.0012.244LCR1
19.631<0.0012.244LCR2
LCT0.8950.90618.867<0.0012.266LCT1
21.855<0.0012.323LCT2
20.139<0.0012.653LCT3
22.120<0.0014.898LCT4
CB0.8720.93918.768<0.0012.484CB1
23.174<0.0012.484CB2
DC0.8760.89822.509<0.0012.245DC1
26.723<0.0012.449DC2
24.300<0.0013.553DC3
CRE1.0001.00020.572<0.0011.000CRE1
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 6. Average variance values of the latent variables.
Table 6. Average variance values of the latent variables.
Latent VariableAverage Variance Extracted
IS0.802
LCR0.872
LCT0.762
CB0.886
DC0.802
CRE1.000
Note: Collusion behavior, CB; Carbon reduction effects; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 7. Correlation coefficients between the square root of the AVE and latent variables (Fornell–Larker).
Table 7. Correlation coefficients between the square root of the AVE and latent variables (Fornell–Larker).
Latent VariableCBCREDCISLCR
CB0.941
CRE0.6321
DC0.6430.6120.895
IS0.6400.6100.6300.896
LCR0.6810.6990.6670.7060.934
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon regimes, LCR.
Table 8. R2, Q2, and cross-validated communality and redundancy.
Table 8. R2, Q2, and cross-validated communality and redundancy.
Latent VariableR2Q2Cross-Validated
Communality		Cross-Validated
Redundancy
CB0.5120.5100.5400.447
CRE0.5660.5431.0000.558
DC0.5170.5140.5750.409
IS 0.576
LCR 0.511
LCT 0.594
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 9. Hypothesis testing results.
Table 9. Hypothesis testing results.
HypothesisRelationshipPath CoefficienttpResult
H1IS → CRE0.0741.4980.134Reject
H2IS → CB0.3176.619<0.001Accept
H3LCT → CRE0.2003.900<0.001Accept
H4LCT → DC0.3868.095<0.001Accept
H5LCR → CRE0.3185.525<0.001Accept
H6LCR → DC0.3948.555<0.001Accept
H7LCR → CB0.4579.997<0.001Accept
H10DC → CRE0.1192.3400.019Accept
H13CB → CRE0.1573.1060.002Accept
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Table 10. Results of the mediation analysis.
Table 10. Results of the mediation analysis.
HypothesisRelationshipPath CoefficienttpResult
H8LCR → DC → CRE0.0472.2120.027Accept
H9LCT → DC→ CRE0.0462.2470.025Accept
H11LCR → CB → CRE0.0722.9520.003Accept
H12IS → CB → CRE0.052.740.006Accept
Note: Collusion behavior, CB; Carbon reduction effects, CRE; Digital control technologies, DC; Industrial structure, IS; Low-carbon technology, LCT; Low-carbon regimes, LCR.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, D.; Ma, C.; Chen, Y.; Wen, A.; Hu, M.; Luo, Q. Carbon Reduction Effects in Transport Infrastructure: The Mediating Roles of Collusive Behavior and Digital Control Technologies. Sustainability 2024, 16, 8390. https://doi.org/10.3390/su16198390

AMA Style

Wang D, Ma C, Chen Y, Wen A, Hu M, Luo Q. Carbon Reduction Effects in Transport Infrastructure: The Mediating Roles of Collusive Behavior and Digital Control Technologies. Sustainability. 2024; 16(19):8390. https://doi.org/10.3390/su16198390

Chicago/Turabian Style

Wang, Da, Chongsen Ma, Yun Chen, Ai Wen, Mengjun Hu, and Qi Luo. 2024. "Carbon Reduction Effects in Transport Infrastructure: The Mediating Roles of Collusive Behavior and Digital Control Technologies" Sustainability 16, no. 19: 8390. https://doi.org/10.3390/su16198390

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop